98 BIRKEI.AND. THE NORWEGIAN AURORA POLARIS EXPEDITION, IQO2 1903. 



the terrella in the above-mentioned wedge-shaped spaces about the poles. The screen here forms in 

 both cases an angle of 270 with its original position (fig. 38 a), and the photographs are now taken from 

 directions that form angles of respectively 120 and 240 with the plane of the screen in its original 

 position, and not, as all the previous ones, from a direction making an angle of 90 with the screen in 

 its original position. 



The way in which the photographs were generally taken was to first expose the plate for about 

 five seconds during the cathode-light experiment, and then, in order to obtain a picture of the terrella 

 itself, to expose the latter for several minutes, illuminated by lamplight. 



These experiments clearly show by analogy how, for instance, cathode rays from the sun will force 

 their way towards the earth in the auroral zone, in such a manner, however, that the bulk of the rays are 

 inclined to slip past it on the night side. The magnetic effect of the rays upon the earth would then 

 be comparable to an ordinary electric current above the earth, whose direction is the reverse of that 

 of the rays, thus approximately from east to west. 



In order to find out whether currents of rays such as these are actually capable of explaining the 

 multiplicity of magnetic perturbations, we must first try to obtain an idea of the exact course of the 

 rays in the vicinity of the earth, and of the relative strength of the bundles of rays. 



Owing to its deflection by terrestrial magnetism, the current from without can, as we have seen, 

 only enter very limited districts, which will alter according as the magnetic axis assumes various 

 positions in relation to the point on the sun that is the source of the rays. 



We must therefore expect to find constant conditions for the current, which, when circumstances 

 are favorable, can force its way down to the earth; at any rate, it will be easy to understand that 

 distinct directions may thereby occur, as the electric rays, in order to come in, must follow paths whose 

 initial direction lies within narrow limits. 



Further, if the rays come from bodies lying outside the earth, the variation in the position of the 

 points of radiation in relation to the magnetic axis, which is occasioned by the rotation of the earth, 

 could give an explanation of the entire movement of the system, as the initial conditions are thereby 

 continually varied. 



If we assume, as, from a physical point of view, we might legitimately do, that the current is of 

 a cosmic nature, and consists of negatively or positively charged corpuscles, the trajectories of the 

 separate corpuscles must, as already stated, more or less approximately follow the magnetic lines of 

 force, moving in spirals about them. 



This will at any rate be the case with the hitherto known rays of this kind, such as ordinary 

 cathode rays, ji rays and a rays, and within a distance from the earth a few times greater than the 

 diameter of the earth. 



We should then, in this perturbation of the isth December, have to consider the effect of a long 

 vertical current, which, in the case of negative corpuscles, must come near to the earth at about Dyra- 

 fjord, or somewhat west of it, answering to an ascending galvanic current. A little above the surface 

 of the earth it turns eastwards, or rather the aggregate effect of the cosmic current relative to the earth 

 is as that of a galvanic current that is directed westwards, or more accurately towards the south-west. 



In this descent of electric corpuscles, some will occasionally come so near the earth that they will be 

 partially absorbed by its atmosphere, and will then eventually give rise to aurora. If the earth were able to 

 retain an electric charge, we should have approximately horizontal currents, which would be necessary 

 for the production of electrical equilibrium. But secondary electric radiation ought also to begin, 

 and then, as it is still influenced by terrestrial magnetism, give rise to vertical ' ray-currents. The 

 bulk of the corpuscles, however, must be imagined, as shown by experimental and theoretical investigations, 

 as able to return, owing solely to this very influence of terrestrial magnetism, and give rise to reversed 



